In the United States, diseases caused by ischemia such as heart disease continue to be a major health problem. One type of disease results in the interruption of nutritional blood flow to the heart, brain or kidney tissue from blockage of a major coronary artery. Although blood flow can often be restored with thrombolytic treatment, it is usually administered too late to prevent irreversible damage. Moreover, the restoration of blood flow also can result in reperfusion injury, where the ultimate degree of tissue damage is greater than would be expected, if the reintroduction of blood and nutrients arrested all damage.
For example, of the 1.5 million people each year who suffer myocardial ischemia (the interruption of coronary blood flow), approximately 700,000 who survive myocardial ischemia that have myocardial infarction (dead heart tissue). For those people who survive with a myocardial infarction, the ischemic damage can form the basis for cardiac arrhythmias, such as bradycardia (abnormally slow heart beats), tachycardia (abnormally fast heart beats) and fibrillation (disorganized heart rhythms wherein the heart quivers rather than beats). About 400,000 people die each year from cardiac arrhythmias.
Protection against infarction has become a long-term goal of cardiology, because infarcted heart muscle cannot be regenerated and is a deficit that the patient must contend with for the remainder of his or her life; therefore a therapy which would cause the heart to better tolerate a period of ischemia is greatly desired. Such a therapy could increase the possibility that timely restoration of the coronary blood flow would salvage myocardial tissue. Such a therapy could also reduce the damage in an area of permanent occlusion. Although many drugs have been proposed to protect the ischemic myocardium, such as beta-blockers, free-radical scavengers, and calcium antagonists, virtually all have performed poorly in whole animal models. Therefore, methods and compounds are needed to both prevent and reduce the damage from ischemic events.
Interestingly, ischemia can be protective as well as injurious and methods for ischemic preconditioning of organs, such as the heart, brain or kidney have been described recently. Ischemic preconditioning refers to a phenomenon whereby a brief period of ischemia renders the myocardium very resistant to infarction from a subsequent schemic insult. [Downey, J. M.: Ischemic preconditioning. Nature's Own Cardioprotective Intervention. TCM 1992; 2:170-176.] The use of this ischemic preconditioning technique consists, in the case of the heart, of interrupting the blood flow through the coronary artery to the heart muscle for five minutes by coronary branch occlusion, followed by reperfusion, or restoring blood flow to the heart, for ten minutes. If the coronary blood flow is restored after five minutes of ischemia, not only will the heart fully recover with no cell death, but the heart will become very resistant to infarction from any subsequent ischemic insult. [See e.g., Liu, Y. and Downey, J. M. Ischemic preconditioning protects against infarction in rat heart. Am J Physiol 1992; 263:H1107-H1112.] Although ischemic preconditioning appears to be universally accepted as a powerful cardioprotectant, it is obviously not the type of intervention that could be administered to the acute myocardial infarction patient.
While the exact mechanism for ischemic preconditioning is not known, as a result of testing in various animal models, ischemic preconditioning appears to be mediated by adenosine which is released during the short ischemic event and populates the adenosine receptors. [Liu, G. S., et al.: Protection against infarction afforded by preconditioning is mediated by A1 adenosine receptors in rabbit heart. Circulation 1991; 84:350-356.] Previously, two types of adenosine receptors have been described, specifically the A1 and A2 adenosine receptors. [Olsson, R. A., Pearson, J. D.: Cardiovascular purinoceptors. Physiol Rev 1990; 70:761-845.] In myocardium, stimulation of A1 adenosine receptors is associated with bradycardia and constricts blood vessels, while A2 receptors mediate coronary vasodilation and inhibit neutrophil activation. Virtually all previous work has indicated that only the A1 adenosine receptor is involved in initiating the preconditioning protection. [See e.g., Downey, J. M., et al., Adenosine and the anti-infarct effects of preconditioning. Cardiovascular Research 1993; 27:3-8.] The A2 adenosine receptor has not been implicated.
As a prospective preconditioning compound, adenosine has been investigated, but the problems associated with administering adenosine outweigh the myocardial benefits. While adenosine does possess ischemic protective abilities, it can only be administered by selective infusion into the coronary artery, as an intravenous injection would cause too much hypotension at doses required for ischemic preconditioning. [Downey, J. M., et al., Adenosine and the anti-infarct effects of preconditioning. Cardiovascular Research 1993; 27:3-8.] Also, increasing evidence exists that adenosine is a mediator of the sensation of anginal pain. Furthermore, adenosine is extremely labile in blood with a half-life of only seconds. [Moser, G. H., et al., Turnover of adenosine in plasma of human and dog. Am J Physiol 1989; 256:C799-C806.] Thus, adenosine is an unlikely candidate for medical treatment of acute myocardial infarction or as a infarction prevention drug.
Adenosine analogues which are capable of selectively activating the A1 adenosine receptor (A1-selective agonists) also have been investigated. Although adenosine analogues which are the A1-selective agonists, such as N.sup.6 -(phenyl-2R-isopropyl)-adenosine (R-PIA) and 2-chloro-N.sup.6 -cyclopentyladenosine (CCPA), have been shown to provide beneficial ischemic preconditioning effects when intercoronarily infused before an ischemic insult, undesirable side effects from activation of the A1 adenosine receptor, e.g. hypotension, A-V conduction delays, bradycardia, narcosis, bronchial spasm, negative inotropic activity and renal vasoconstriction, may present an insurmountable obstacle to achieving a practical therapy based on parenteral administration of A1-selective agonists. [See e.g., Thornton, J. D., et al., Intravenous Pretreatment With A1-Selective Adenosine Analogues Protects the Heart Against Infarction. Circulation, 1992; 85:659-665.]
Recently, a third adenosine receptor, the A3, has been characterized. [Zhou, Q. Y., et al., Molecular cloning and characterization of an adenosine receptor: The A3 adenosine receptor. Proc Natl Acad Sci 1992; 89:7432-7436.] It is reported that the A3 adenosine receptor cloned from rat brain, designated R226, shares high sequence identity with the two previously identified adenosine receptors. R226 has been reported to bind the non-selective adenosine agonists N-ethyladenosine 5'-uronic acid (NECA), and the A1-selective agonist N.sup.6 -2-(4-amino-3-iodophenyl)-ethyladenosine (I-APNEA), but not to bind the A1-selective antagonists 1,3 dipropyl-8-cyclopentylxanthine (DPCPX) and 8-{4-[({[(2-aminoethyl)amino]carbonyl}methyl)oxyl-phenyl}-1,3-dipropylxant hine (XAC). Nothing has been currently reported about its physiological role.
Firestein, et al., U.S. patent application Ser. No. 08/070,689, now abandoned, describes DNA encoding Human adenosine A3 receptor, isolated human A3 receptor, cells expressing the same and uses thereof.
Andrew J. van Bergen, et al., Abstracts of Papers, 206th National Meeting of the American Chemical Society, Chicago, Ill., 1993, MEDI 217, proposed a preliminary structure activity relationship for some A3 receptor agents. Further, Kenneth A. Jacobson, et al., Abstracts of Papers, 207th National Meeting of the American Chemical Society, San Diego, Calif., 1994, MEDI 244, reported N6-benzyladenosine 5'-uronamides as selective agonists for rat brain A3 receptor. This work was also published in expanded form as "Structure-Activity Relationships of N.sup.6 -Benzyladenosine-5'-uronamides as A3-Selective Adenosine Agonists" in J. Med. Chem., 37, 636 (1994).
Christopher Salavatore, et al., Proc. Natl. Acad. Sci. USA 90 (1993) reported the binding constants for several agents to the human A3 adenosine receptor.
Joel Link, et al., Mol. Pharmacol. 44:524-532 (1993) also proposed a preliminary structure activity relationship for binding of some A3 receptor agents to recombinant sheep A3 adenosine receptor.